SolanaLM Documentation¶
Decentralized AI Network on Solana
Under active development
SolanaLM is under active development. APIs, schemas, and on-chain layouts may change. Production use at your own risk. Issues + PRs welcome — see the GitHub repo.
SolanaLM is a hybrid decentralized network combining LLM inference and federated learning on Solana. Nodes earn SOL through dual revenue streams: serving inference requests and participating in training rounds.
Key Features¶
-
:robot:{ .lg .middle } Multi-Backend AI Inference
Run local models with PyTorch/Transformers or connect to external APIs like OpenAI, Anthropic, and Cohere through a unified interface.
-
:brain:{ .lg .middle } Federated Learning
State-of-the-art algorithms (FedAvg, FedProx, SCAFFOLD) with differential privacy and secure aggregation.
-
:lock:{ .lg .middle } Privacy-Preserving
Onion routing, anonymous payments, and encrypted inference for sensitive workloads.
-
:coin:{ .lg .middle } Earn SOL
Dual revenue streams from inference requests and training participation with dynamic pricing.
Quick Start¶
Get up and running in minutes:
# Clone and setup
git clone https://github.com/cryptuon/solanalm.git
cd solanalm && poetry install && poetry shell
# Start development environment
python scripts/quick_start.py
# Test the system
python scripts/test_end_to_end.py
:octicons-arrow-right-24: Full Installation Guide
Basic Usage¶
Python SDK¶
import asyncio
from solanalm_client import SolanaLMClient
async def main():
async with SolanaLMClient("http://localhost:8001") as client:
response = await client.inference(
model="microsoft/DialoGPT-small",
prompt="Explain quantum computing",
wallet_address="your-wallet-address"
)
print(f"Response: {response.response}")
print(f"Cost: {response.cost_sol} SOL")
asyncio.run(main())
OpenAI-Compatible API¶
from solanalm_client import OpenAICompatibleClient
client = OpenAICompatibleClient(
base_url="http://localhost:8001/v1",
api_key="your-solana-wallet-address"
)
response = client.chat.completions.create(
model="microsoft/DialoGPT-small",
messages=[{"role": "user", "content": "Hello!"}]
)
:octicons-arrow-right-24: SDK Documentation
Architecture Overview¶
┌─────────────────────────────────────────────────────────────────┐
│ User Interfaces │
│ ┌───────────┐ ┌───────────┐ ┌───────────┐ ┌─────────────┐ │
│ │Python SDK │ │ OpenAI API│ │Web Dashboard│ │Terminal TUI│ │
│ └─────┬─────┘ └─────┬─────┘ └─────┬─────┘ └──────┬──────┘ │
└────────┼──────────────┼──────────────┼───────────────┼──────────┘
└──────────────┴──────────────┴───────────────┘
│
┌──────────▼──────────┐
│ Gateway API │
│ Load Balancer │
└──────────┬──────────┘
│
┌──────────▼──────────┐
│ Node Registry │
└──────────┬──────────┘
│
┌───────────────────────┼───────────────────────┐
│ │ │
┌──────▼───────┐ ┌──────────▼─────────┐ ┌───────▼───────┐
│ Inference │ │ Training Nodes │ │ Proxy Nodes │
│ Nodes │ │ • Federated FL │ │ • OpenAI API │
│ • PyTorch │ │ • Privacy FL │ │ • Anthropic │
│ • llama.cpp │ │ • Model Training │ │ • Cohere │
└──────────────┘ └────────────────────┘ └───────────────┘
:octicons-arrow-right-24: Architecture Details
Deployment Options¶
| Environment | Description | Use Case |
|---|---|---|
| Local | Single-command setup | Development, testing |
| Docker | Full containerized stack | Staging, small deployments |
| Kubernetes | Production-grade orchestration | Enterprise, high availability |
:octicons-arrow-right-24: Deployment Guide
Community & Support¶
- GitHub Issues: Report bugs and request features
- GitHub Discussions: Ask questions and share ideas
- Marketing site: solanalm.cryptuon.com
License¶
SolanaLM is open-source software licensed under the MIT License.